Evaluation of enzyme-linked immunoassay for serological diagnosis of cysticercosis.

We evaluated a commercially available enzyme-linked immunoassay (ELISA) from LMD Laboratories, Inc., Carlsbad, Calif., for the detection of antibodies in serum to the cysticercus of Taenia solium. The ELISA was performed on 308 serum samples; 198 from a pool of healthy individuals, 42 from patients who had antibodies against a variety of parasites other than T. solium, and 68 from patients suspected of having cysticercosis. All of these 68 specimens were tested both by the ELISA and by an immunoblot method (enzyme-linked immunoelectrotransfer blot assay [EITB]) developed at the Parasitic Serology Laboratory of the Centers for Disease Control and Prevention. Twenty-seven of the 68 serum samples from patients suspected of having cysticercosis were positive by both EITB and ELISA, while 31 were negative by both assays. ELISA results for three and two samples were considered false positive and false negative, respectively, when compared with the results of EITB. Results for an additional five samples were considered equivocal but were technically positive because their optical density readings were slightly above the cutoff value. Three of the 198 serum samples from the bank of serum samples from healthy individuals were also false positive by ELISA (the EITB result for the samples was negative). Six other serum samples from healthy individuals which had equivocal results and the five serum samples from patients with equivocal results were EITB negative. Serum samples containing antibodies against Echinococcus spp. frequently cross-reacted with the cysticercus ELISA antigen (13 of 16 specimens), but serum samples with antibodies against other parasites did not (2 of 26 specimens); all of these serum samples were EITB negative. The commercially available ELISA that we describe is a simple and rapid test. Considering all 308 specimens, the ELISA had a specificity of 93% (when samples with equivocal results were considered negative) or 89% (when samples with equivocal results were considered positive); the sensitivity was 93%.